DESCRIPTION (adapted from the Abstract): In the United States, infection with Mycobacterium avium is the most common disease of bacterial origin among patients with the acquired immunodeficiency syndrome (AIDS). Despite the great morbidity and mortality associated with M. avium infections, we lack a detailed understanding of the epidemiology of this pathogen. For instance, we do not yet know whether an animal reservoir exists of M. avium clones that cause disease in AIDS patients; we have much to learn about the transmission of the bacterium during the course of infection. In this application, the Principal Investigator proposes studies that will (a) provide a comprehensive understanding of the molecular epidemiology of M. avium infections in humans (primarily, in AIDS patients) and animals, and (b) enable the development of powerful new genetic markers for strain identification. Preliminary studies have led the Investigator to formulate three hypotheses: (i) the majority of M. avium infections in patients with AIDS are caused by only a small subset of extant M. avium clones; (ii) an animal reservoir exists of M. avium clones that may also be recovered from AIDS patients; and (iii) automated polymerase chain reaction (PCR)-based DNA fingerprinting strategies may be used for the rapid and unambiguous identification of M. avium clones. To test these hypotheses, the Investigator and his associates will use multilocus enzyme electrophoresis--the method of choice for bacterial population genetic studies--to determine allelic variation at metabolic enzyme loci in a sample of more that 1,000 M. avium isolates recovered from humans (primarily, from AIDS patients) and from domesticated and wild animals. All isolates in the collection will also be characterized by IS1245 fingerprinting, a technique that will provide a fine-structure epidemiologic analysis of M. avium from humans and animals. Finally, because present methodology for M. avium strain identification and fingerprinting are generally labor intensive and time consuming, the researchers will evaluate the feasibility of used a semi-automated PCR-based method (i.e., flourophore-enhanced repetitive-element PCR) for M. avium strain identification.